Intrinsically photosensitive retinal ganglion cells (ipRGCs) express the photopigment melanopsin. They were once viewed as a single cell type conveying light-intensity (irradiance) information to 'non-image-forming' (NIF) centers for circadian, pupillary and neuroendocrine regulation. However, ipRGCs have proved to be diverse, with some subtypes contributing to cortical visual mechanisms. Understanding this diversity in relation to visual behavior and perception is major long-term goal of this project. Ou first aim is to develop detailed structural and functional descriptions of two novel ipRGC types, the M5 and M6 cells. Both have small fields and unusual functional properties and may contribute to cortical vision. M5 cells may be spectrally opponent, while M6 cells exhibit variable and surprising features such as transient or suppressed-by-contrast responses. Fulfilling this aim will largely complete the inventory of ipRGC types and advance a full accounting of ganglion cell types in the mouse, now the premiere model for retinal studies. Though ipRGCs are diverse, most of them share a striking property - the ability to report environmental light levels. This suits the requirement of many NIF reflexes for a signal encoding global light intensity, rather than spatial contrast, color or movement. Our second aim is to reveal the mechanistic basis of this irradiance-coding capacity. What are the cells, circuits, and synaptic mechanisms that permit ipRGCs to encode irradiance, while conventional ganglion cells cannot? We will focus on the bipolar cells that link rod and cone photoreceptors to ipRGCs, asking how these convey raw irradiance information to the ipRGCs while bipolars that feed conventional RGCs filter out such information to encode mainly changes in intensity (spatiotemporal luminance contrast). Dendrites of ipRGCs are segregated in two specific sublaminae of the inner plexiform layer, comprising only part of the ON sublayer. This implies that they receive inputs from subsets of ON bipolar cells specialized for irradiance coding. Differences in inhibitory circuits also appear likely to contribute. We propose to determine which ipRGC types actually encode irradiance, to trace the transmission of irradiance information through specific presynaptic bipolar types and synapses, and to contrast these with the bipolar circuits driving RGCs than cannot report irradiance. We will assess the relative contributions of outer-retinal synaptic mechanisms, glutamatergic signaling from bipolar to ganglion cells, and amacrine-cell inhibition to the capacity of ON bipolar outputs to encode irradiance. To answer these questions, we will augment our established technical arsenal (mouse genetic models, patch clamp recording, light stimulation, pharmacological manipulation, intracellular dye-filling, and confocal microscopy), with powerful new tools, including viral transsynaptic tracing, two-photon microscopy and an exciting new imaging technique for monitoring glutamate release from bipolar terminals at specific depths in the IPL onto identified ipRGCs.